Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Neurotrophins (NTs) are a family of related polypeptide growth factors, including nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), NT-3, NT-4/5, and NT-6 (1). These molecules bind two types of cell surface receptors characterized by different binding affinities and molecular weight, including a 75-kD low-affinity glycoprotein receptor (p75) and 140- to 145-kD high-affinity receptors recognized by a tyrosine-specific protein kinase (Trk) activity. The physiologic role of NTs in promoting differentiation and survival of developing neurons in the central and peripheral nervous system has been clearly delineated (2). Moreover, NTs stimulate differentiation and proliferation of several cell types of all three germ layers (3, 4).

NTs and NGF play a role in the modulation of certain human malignancies, including those of neurogenic and ectodermal origin (5, 6). They are also involved in stimulation of clonal growth on human lung cancer cells in vitro via high-affinity NT receptors (5, 7, 8). However, studies on the effects of NTs on neoplastic cell lines often reported conflicting results. NTs can induce differentiation without having an effect on cell growth in some tumor cells, whereas inhibition of neoplastic cell proliferation has been correlated with translocation of NGF to the cell nucleus and binding to a chromatin receptor (9). Some tumors were nonresponsive to NGF in vitro, whereas NGF has been shown to be mitogenic for cultured cells from medullary thyroid carcinoma, and breast and prostate cancer cells (13, 14). The proliferation of human lung cancer is regulated by several growth factors both in vivo and in vitro (15) via autocrine and/or paracrine mechanisms. Growth factors are now considered to play a crucial role in lung cancer cell proliferation (19, 20).

In the past few years, efforts were made to characterize and discover factors able to slow down or arrest the neoplastic progression. Growth factors are obvious candidates for this based on their ability to control growth and differentiation.

Although NTs seem to play a role in growth and differentiation of several human malignancies, their role in lung tumors has received little attention. The present study was therefore designed to investigate the expression and distribution of NTs and high- and low-affinity NT receptors in different human lung malignancies by Western blot analysis and immunohistochemistry. NT and NT receptor expression was linked to tumor proliferative activity, using Ki-67 immunohistochemistry as a marker of tumor proliferation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects and Tissue Preparation

Surgical samples of tumors of the lower respiratory tract were removed from 38 patients (26 men and 12 women, 54 to 75 yr of age) undergoing surgical treatment. Twelve well-differentiated acinar adenocarcinomas, eight bronchioloalveolar carcinomas, 10 well-differentiated squamous cell carcinomas, and eight small cell lung cancers (SCLC) were studied.

Specimens were dissected, necrotic areas were discarded, and samples were fixed and further processed for histology and immunohistochemistry or homogenized and used for Western blot analysis.

Antisera

For the immunohistochemical detection of NTs and NT receptors, the following antisera were used. (1) Rabbit polyclonal TrkA immunoglobulin recognizes an epitope corresponding to amino acids 763 to 777, mapping adjacent to the carboxy terminus of human trkA p140, and is not cross-reactive with TrkB or TrkC. (2) Rabbit polyclonal TrkB immunoglobulin, according to the manufacturer's instructions, recognizes an epitope corresponding to amino acids 794 to 808 of mouse TrkB p145 and is not cross-reactive with TrkA or TrkC. (3) Rabbit polyclonal TrkB [TK-] immunoglobulin recognizes the carboxy terminus of the truncated [TK-] TrkB protein precursor gp95 of mouse originspecific recognition of mouse, rat, and human TrkB [TK-] gp95and is not cross-reactive with TrkB gp145, TrkA gp140, or TrkC gp140. (4) Rabbit polyclonal TrkC immunoglobulin recognizes an epitope corresponding to amino acids 798 to 812 of porcine TrkC p140 and is not cross-reactive with TrkA or TrkB. (5) Goat polyclonal antibody to human p75NT receptor was also used. (6) Rabbit anti-NGF polyclonal antibody displays less than 1% cross-reactivity against recombinant human NT-3, NT-4, and BDNF. (7) Rabbit polyclonal antibody anti-BDNF corresponds to the amino terminal of mouse BDNF coupled to ovalbumin and does not cross-react with NT-3 or NGF. (8) Rabbit polyclonal antibody anti NT-3 was raised against the amino terminal of mouse NT-3 coupled to bovine serum albumin (BSA) and does not cross-react with BDNF or NGF. It recognizes the amino-acid sequence mapping the carboxy terminus of the p75 NT receptor precursor of human origin and is not cross-reactive with other growth factor receptors. (9) Rabbit polyclonal antibody anti-Ki-67 recognizes the amino-acid sequence 2597-2896, mapping at the carboxy terminus of Ki-67 of human origin specific for Ki-67.

The specificity of antibodies for corresponding peptides was assessed by Western blot analysis, using homogenates of rat brain as NT and NT receptor source, with the exception of human -NGF used as a standard in this control experiment.

Western Blot Analysis

Homogenates of lung tumors were centrifuged at 1,500 × g to remove nuclei and cell debris. The supernatant was resuspended in an immunoprecipitation assay buffer containing phenylmethylsulfonylfluoride, aprotinin, and leupeptin. Aliquots of supernatant were used for protein assay against a standard of BSA. Defined amounts (50 µg) of proteins were loaded onto 10% stacking sodium dodecyl sulfate-polyacrylamide gel and electrophoresed through a 10% sodium dodecyl sulfate-polyacrylamide gel. After electrophoresis, proteins were transferred to nitrocellulose paper. Standard human -NGF (10 ng) and rat brain proteins were used as a reference for the NTs and NT receptors. Antibodies were dissolved in 0.1 M phosphate-buffered saline (PBS) containing BSA (1%) and Tween 20 (0.05%). Optimal antibody concentrations were established in a series of preliminary experiments. The specificity of immune reaction was assessed using antibodies preadsorbed with corresponding peptides. Anti-NGF, anti-BDNF, and anti-NT-3 antibodies (dilution 1:3,000) and anti-TrkA, anti-TrkB (both full and truncated isoforms), anti-TrkC (dilution 1:500), and anti-p75NT (dilution 1:50) receptor antibodies were then applied.

A secondary horseradish peroxidase-conjugated antibody was used in PBS containing nonfat milk (5%) and Tween 20. The reaction was detected using a specific Western blotting detection reagent (ECLTM RPN 2106; Amersham Pharmacia) and developed using a chemiluminescence Hyperfilm (Amersham).

Immunohistochemistry

Serial 5-µm-thick sections were obtained from formalin-fixed tissues. Paraffin-embedded specimens were cut using a rotatory microtome. Sections were mounted on gelatin-coated slides and processed for immunohistochemistry as described elsewhere (21). Briefly, from each paraffin block, consecutive sections were exposed to anti-TrkA, anti-TrkB, anti-TrkC (dilution 1:100), and anti-p75 antibodies (dilution 1:10) alone or in the presence of the antibodies preadsorbed with corresponding peptides (10 µg/ml), and to anti-NGF, anti-BDNF, and anti NT-3 antibodies (dilution 1:1,000) and to the antibodies preadsorbed with human -NGF (10 µg/ml), human BDNF blocking peptide (10 µg/ml), and human NT-3 (10 µg/ml).

Analysis of cell proliferation was determined using an antibody against the nuclear protein Ki-67. This antibody is widely used to determine Ki-67 protein that is expressed in proliferating cells and may be required for maintaining cell proliferation (dilution 1:400). Optimal antisera dilutions and incubation times were assessed in a series of preliminary experiments. After incubation, slides were rinsed twice in phosphate buffer and exposed for 30 min at 25°C to antirabbit (for Trk, NT, and Ki-67 immunohistochemistry) and antigoat (for p75NT receptor immunohistochemistry) secondary antibodies at a dilution of 1:100. The product of immune reaction was revealed using 0.05% 3,3-diaminobenzidine in 0.1% H₂O₂ as a chromogen. Sections were then washed, dehydrated in ethanol, mounted in a synthetic mounting medium, and viewed with a light microscope. Endogenous peroxidase activity was blocked by H₂O₂, whereas the nonspecific binding of immunoglobulin to glass and tissue was prevented by 3% fetal calf serum added to the incubation medium. In a series of preliminary experiments, immunohistochemistry was performed using both paraffin-embedded and frozen sections. No differences in the intensity or distribution of immunostaining were noticeable using the two types of sections, although microanatomical details were better preserved in paraffin-embedded material (data not shown). In view of this, similarly as reported by other investigators (22), paraffin-embedded material was used in standard immunohistochemistry experiments.

Quantitative Image Analysis

The intensity of the immunoreaction developed within the cell body of neoplastic cells was assessed microdensitometrically by a program of an IAS 2000 image analyzer (Delta Sistemi, Rome, Italy) connected via a TV camera to the microscope. Sections were examined at a final ×200 magnification. The system was calibrated, taking as zero the background obtained in sections exposed to preimmune serum. Ten 100-µm² areas were delineated in each section by a measuring diaphragm. Sections processed for Ki-67 immunoreactivity were examined independently at a final ×450 magnification by two researchers to assess proliferation score. A high proliferation score was given when > 25% cells per microscope field displayed Ki67 immunoreactivity. A low proliferation score was given when < 25% cells per microscope field displayed Ki-67 immunoreactivity.

Statistics

Data of quantitative analysis of the intensity of the immune staining for NTs and NT receptors in the tissues examined were analyzed statistically by analysis of variance followed by Duncan's multiple range test as a post hoc test.

Chemicals

Anti-NGF, BDNF, NT-3, the corresponding blocking peptides human -NGF, human BDNF, human NT-3; anti-TrkA, anti-TrkB full-length, anti-TrkB truncated, and anti-TrkC antibodies; peptides used for rising anti-TrkA, TrkB full-length and truncated isoforms, TrkC antibodies, the antibody against p75NT receptor, the corresponding receptor blocking peptide, and the antibody against Ki-67 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Horseradish peroxidase-conjugated secondary antibodies for Western blot analysis, antigoat and antirabbit were also purchased from Santa Cruz Biotechnology. Antigoat peroxidase-conjugated secondary antibody was purchased from Sigma Chemical Co. (St. Louis, MO). Antirabbit IgG peroxidase-conjugated secondary antibody was obtained from Boehringer Mannheim GmbH (Mannheim, Germany). Other chemicals were obtained from Sigma Chemical Co. or Merck (Darmstadt, Germany).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Western Blot Analysis

The first step was to evaluate the specificity of anti-NT or anti-NT receptor antibodies determined by Western blot analysis. The results of immunoblot analysis of antibodies tested against human -NGF and homogenates of rat brain expressing NTs and the NT receptors are shown in Figures 1-3 (lane 1). NGF antibody was bound to a single band at approximately 14 kD (Figure 1A, lane 1). BDNF antibody reacted with a single band at approximately 14 kD (Figure 2A, lane 1). NT-3 antibody reacted with a single band of approximately 14 kD (Figure 3A, lane 1). The TrkA, TrkB (full-length), and TrkC antibodies reacted with a single band of 140 kD (Figures 1B, 2B, and 3B, respectively, lane 1). The truncated isoform of TrkB receptor reacted with a band at approximately 95 kD (Figure 2C, lane 1). p75NT receptor antibody reacted with a single band of 75 kD (Figure 1C, lane 1). Preadsorption of antibodies with corresponding NT peptides or NT receptor blocking peptides caused the disappearance of immune bands (Figures 1-3 [lane 2], 4A [lanes 2 and 4], and 4B [lanes 2, 3, 5, and 6]). Immunoblots of homogenates from human tumors displayed a binding pattern of anti-NGF, BDNF, and NT-3 antibodies (Figures 1-3, lanes 3-6) as well as of anti-TrkA, TrkB, TrkC, and p75NT receptor antibodies (Figures 1-3, lanes 3-6), similar to that generated by human -NGF and proteins from homogenates of rat brain.

View larger version (71K): [in this window] [in a new window]   Figure 1.   Western blot analysis of NGF (A), TrkA (B), and p75 (C ) antibodies. Lane 1: Immunoblots with human -NGF and membranes obtained from rat brain exposed to different antibodies tested. Lane 2: Immunoblots with human -NGF and membranes obtained from rat brain exposed to the different tested antibodies preabsorbed with the corresponding receptor blocking peptides. The antibodies preabsorbed with peptides did not develop a band of immunoreactivity. Lanes 3-6: Immunoblots with membranes obtained from human bronchioloalveolar carcinomas, adenocarcinomas, squamous cell carcinomas, and SCLC specimens, respectively.

View larger version (59K): [in this window] [in a new window]   Figure 2.   Western blot analysis of BDNF (A), TrkB full-length (B), and TrkB truncated (C ) antibodies. Lane 1: Immunoblot analysis with membranes obtained from rat brain exposed to the different antibodies tested. Lane 2: Immunoblot analysis with membranes obtained from rat brain exposed to the different tested antibodies preabsorbed with corresponding receptor blocking peptides. The antibodies preabsorbed with peptides did not develop a band of immunoreactivity. Lanes 3-6: Immunoblot analysis with membranes obtained from human bronchioloalveolar carcinoma, adenocarcinoma, squamous cell carcinoma, and SCLC specimens, respectively.

View larger version (62K): [in this window] [in a new window]   Figure 3.   Western blot analysis of NT-3 (A) and TrkC (B) antibodies. Lane 1: Immunoblot analysis with human membranes obtained from rat brain exposed to the different antibodies tested. Lane 2: Immunoblot analysis with membranes obtained from rat brain exposed to the different tested antibodies preabsorbed with corresponding receptor blocking peptides. The antibodies preabsorbed with peptides did not develop a band of immunoreactivity. Lanes 3-6: Immunoblot analysis with membranes obtained from human bronchioloalveolar carcinoma, adenocarcinoma, squamous cell carcinoma, and SCLC specimens, respectively.

A migration band corresponding to NGF protein was demonstrated in bronchioloalveolar and squamous cell carcinomas (Figure 1A, lanes 3 and 5), whereas an inconstant, faint band was demonstrated in adenocarcinomas and SCLC samples (Figure 1A, lanes 4 and 6). A band corresponding to BDNF protein was detected in SCLC (Figure 2A, lane 6). Less intense bands were observed using samples obtained from bronchioloalveolar carcinomas, adeno-carcinomas, and squamous cell carcinomas (Figure 2A, lanes 3, 4, and 5). A clear band corresponding to NT-3 protein was found in squamous cell carcinomas only (Figure 3A, lane 5).

A band corresponding to TrkA protein was documented in adenocarcinomas and squamous cell carcinomas (Figure 1B, lanes 4 and 5), whereas a less developed band was detected for bronchioloalveolar carcinomas and SCLC (Figure 1B, lanes 3 and 6). A strong band corresponding to TrkB full-length and a less intense band corresponding to the truncated isoform proteins were detected primarily in SCLC (Figures 2B and 2C, lane 6). A less intense and inconstant band was also observed in bronchioloalveolar carcinoma, adenocarcinoma, and squamous cell carcinoma samples (Figures 2B and 2C, lanes 3, 4, and 5). A band corresponding to TrkC protein was observed in squamous cell carcinomas (Figure 3B, lane 5), whereas a faint band was detected in SCLC (Figure 3B, lane 6). No p75 proteins were found in our samples (Figure 1C).

In a series of control experiments, protein homogenates derived from lung neoplasms were incubated with antibodies preabsorbed with corresponding peptides or preincubated with NGF, BDNF, or NT-3 before electrophoresis and analysis of Trk receptors (Figures 4A and 4B). This preincubation caused the disappearance of bands of immunoreactivity (Figures 4A [lanes 2 and 4] and 4B [lanes 2, 3, 5, and 6]), suggesting that NTs bind the relative Trk receptors.

View larger version (18K): [in this window] [in a new window]   Figure 4.   Western blot analysis of NT (A) and Trk (B) antibodies. (A) NTs. Immunoblots with membranes obtained from human bronchioloalveolar and squamous cell carcinomas exposed to NGF antibody (lanes 1 and 3), with membranes obtained from human bronchioloalveolar carcinomas and SCLCs (lanes 1 and 3) exposed to BDNF antibody, or with membranes obtained from human squamous cell carcinomas (lane 1 ) exposed to NT-3 antibody and with the antibodies preabsorbed with the corresponding receptor blocking peptides (lanes 2 and 4 ). Preabsorption of antibodies with peptides did not result in the development of bands of immunoreactivity. (B) Trks. Immunoblots with membranes obtained from human bronchioloalveolar and squamous cell carcinomas exposed to TrkA receptor antibody (lanes 1 and 4 ), with membranes obtained from human small cell carcinoma and bronchioloalveolar carcinoma exposed to TrkB receptor antibody (lanes 1 and 4 ), or with membranes obtained from human squamous cell carcinoma and SCLC exposed to TrkC receptor antibody (lanes 1 and 4 ). The antibodies preabsorbed with corresponding blocking peptides (lanes 2 and 5 ) as well as the membranes preincubated with corresponding ligands, NGF, BDNF, or NT-3 respectively (lanes 3 and 6 ), did not develop or developed a faint band of immunoreactivity.

NT and NT Receptor Immunohistochemistry

Sections of lung cancer exposed to NT or NT receptor antibodies developed a dark-brown (intense) or yellow-brown (slight) immunostaining or no staining (Figures 5-7). No immunostaining developed in sections incubated with antibodies preadsorbed with the peptides used for raising them (Figures 5-7) or with a preimmune serum. Immunostaining was located in cancer cells, stromal fibroblasts, immune cells, and vessels surrounding tumors. NT and NT receptor expression within different tumor cell types is summarized in Tables 1 and 2, respectively.

View larger version (121K): [in this window] [in a new window]   Figure 5.   Micrographs of NGF and TrkA immunostaining in consecutive sections of human bronchioloalveolar carcinomas (A, B) and adenocarcinomas (D, E ) of the lower respiratory tract. (C, F ) Nonspecific immunostaining obtained when the sections were exposed to the NGF antibody preabsorbed with the corresponding blocking peptide. Note the localization of specific immunostaining within tumor cells and within the stroma surrounding the tumor.

View larger version (172K): [in this window] [in a new window]   Figure 6.   BDNF (A), TrkB (B), and NT-3 (C ) immunostaining in sections of human squamous cell carcinomas of the lower respiratory tract. (D) Nonspecific immunostaining obtained when the sections were exposed to BDNF antibody preabsorbed with the corresponding blocking peptide. Note the specific immunostaining located within the tumor cells.

View larger version (52K): [in this window] [in a new window]   Figure 7.   BDNF (A) and TrkB (B) immunostaining in consecutive sections of human small cell carcinomas of the lower respiratory tract. (C ) Nonspecific immunostaining obtained when the sections were exposed to the BDNF antibody preabsorbed with the corresponding blocking peptide. Note the localization of specific immunostaining within the tumor cells.

NT and NT receptor immunoreactivity was also detected in uninvolved lung cells. NGF immunoreactivity was observed in mucous glands, artery and vein walls, nerve fibers, stromal cells, lymphocytes, and both alveolar and interstitial macrophages. BDNF and NT-3 immunoreactivity was demonstrated in vascular structures and in nerve fibers (data not shown). Trk immunoreactivity was detected in arteries and veins, nerve fibers, fibroblasts, lymphocytes, and both alveolar and interstitial macrophages (data not shown).

Approximately 28% of highly differentiated adenocarcinomas developed a weak and inconstant NGF and BDNF immunoreactivity within tumor cells (Table 1). Approximately 33% of bronchioloalveolar carcinomas developed an intense immunostaining within the cytoplasm of neoplastic cells, within stromal fibroblasts, and in arteries and veins surrounding neoplasms (Table 1). A small percentage of SCLCs developed a faint NGF and BDNF immunoreactivity within tumor cells (Table 1). About 21% of highly differentiated squamous cell carcinomas developed a faint BDNF immunoreactivity and 16% an intense NT-3 immunoreactivity within the cytoplasm of large-size tumor cells (Table 1 and Figures 5-7).

Approximately 46 and 30% of neoplastic cells of adenocarcinomas developed a cytoplasmic immunostaining of differing intensity for TrkA and TrkB, respectively (Table 2). The immunoreaction was also demonstrated within stromal fibroblasts as well as in vessels surrounding neoplastic lesions (data not shown). A total of 42% of bronchioloalveolar carcinomas developed an intense TrkA immunostaining but not TrkB and TrkC immunostaining (Table 2). Approximately 24 and 35% of well-differentiated squamous cell carcinomas displayed an intense TrkB (for both full-length and truncated isoforms) and a faint TrkC cytoplasmic immunostaining, respectively (Table 2). In squamous cell carcinomas, TrkA immunoreactivity was found within stromal fibroblasts and nerve fibers as well as in vessels surrounding neoplastic lesions but not in the tumor cells (Table 2). A faint and inconstant TrkA and TrkC immunoreaction was observed in 18 and 55% of SCLCs, respectively, whereas 65% of them developed an intense TrkB immunoreactivity (Table 2 and Figures 5-7). No p75 receptor immunoreactivity was observed in the cells of the different lung tumors investigated. Sparse nerve fibers supplying the lung parenchyma displayed a p75 immunoreactivity (data not shown).

Ki-67 Immunohistochemistry

Ki-67 antibody generated a nuclear immunoreactivity abundant in poorly differentiated zones and absent in the best differentiated areas. High Ki-67 scores were observed in neoplasm sections not expressing NT and NT receptor immunoreactivity (Figure 8).

View larger version (110K): [in this window] [in a new window]   Figure 8.   Micrographs of Ki-67 and NT and NT receptor immunoreactivity in adenocarcinoma (A, B), squamous cell carcinoma (C, D), bronchiolo-alveolar carcinoma (E, F ), and SCLC (G, H ). A nuclear staining was clearly noticeable in sections exposed to Ki-67 antibody (A, C, E, and G ). No immunoreactivity was noticeable in consecutive sections (B, D, F, and H ) exposed to NT and/or NT receptor antibodies.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our study reports for the first time the expression and the anatomic localization of NTs and NT receptor proteins in different types of malignant tumors of the lower respiratory tract, characterized by Western blotting and immunohistochemistry. The role of NTs in lung tumors was only sparsely investigated (5, 22). Previous immunohistochemical data revealed that TrkA receptor immunoreactivity occurred in about 54% of adenocarcinomas and in about 80% of squamous cell carcinomas of the lung (22), whereas in vitro functional studies demonstrated that lung cancer cells were responsive to NGF via high-affinity binding sites (5). We found that human non-SCLC and SCLC express NTs as well as high-affinity NT membrane receptors.

Lung neoplasms express NTs and NT receptors despite the fact that normal counterpart cells are not immunoreactive to these peptides. It is reasonable to hypothesize that malignant transformation may be linked to functional and receptor expression changes because of modified growth and differentiation necessity. Otherwise, it cannot be dismissed that NTs may have a function in the initiation and progression of lung tumors. Although colocalization of NTs and NT receptors was not performed, the analysis of immunoreaction in consecutive sections revealed the possible coexistence of NTs and their related receptors within the same cells. These data support the hypothesis of an autocrine mechanism operating in these tissues. Moreover, the expression of NTs in stromal fibroblast and immune cells (macrophages and lymphocytes) surrounding tumor cells indicates that these structures may be a source of NTs outside the tumor, acting on tumor cells expressing NT receptors via a paracrine mechanism. These findings provide further evidence for a possible extratumor release of NGF and the occurrence of a paracrine mechanism as in esophageal, breast, and prostate cancer tissues (22).

In our experimental conditions, we did not demonstrate low-affinity p75 receptor immunoreactivity. The occurrence of TrkA but not p75 receptor immunoreactivity suggests that TrkA alone may modulate biologic activity of NGF in lung neoplasms. This hypothesis is consistent with data of responses to NGF by nerve and other cells in the presence of blocking antibodies to p75 or in cells lacking low-affinity NT receptors (5, 13, 25, 26). It was also reported that several TrkA immunoreactive neuronal and non-neuronal cancers did not express the p75 receptor (22, 27). Recently, it has been shown that high TrkA/p75 ratio and/or TrkA receptor expression in the absence of p75 carry out a mitogenic function in a pancreatic carcinoid cell line (30, 31), although the absence of the p75 receptor is reported in metastatic cells (28, 29).

Although the role of NTs in the control of the growth of pulmonary non-SCLC and SCLC is not yet clarified, it is possible to hypothesize that NTs may influence several tumor functions. The inverse relationship observed between a high proliferation Ki-67 score and NT and NT receptor immunoreactivity suggests that NTs play a functional role in modulating differentiation and growth of lung tumors. Therefore, as described for several human non-neuronal tumors, NFG, BDNF, and NT-3 may control tumor cell division and proliferation (5). It cannot be dismissed that they may facilitate basement membrane penetration by inducing proteolytic enzyme release (32) as well as malignancy progression toward the brain, allowing tumor cell penetration into the matrix of the NT-rich stromal brain microenvironment (22, 35).

In conclusion, our data support the hypothesis that NTs may be operative in modulating growth and differentiation of human lung tumor cells possessing high-affinity receptors for such molecules, yielding new perspectives in understanding mechanisms that regulate tumor cell behavior.